Predictive Analytics can help significantly in production operations, but also in the streamlining of the supply chain. It is already bringing results to the leading logistics operators.
Logistics companies implement predictive analytics to streamline supply chains, processes and operations. But not only that. Thanks to the collected and thoroughly analysed data, logistics specialists can prevent equipment failures by extending its failure-free operation.
Danish maritime carrier Maersk Line, which operates in more than 130 countries (has more than 600 container ships), annually transports goods worth almost $700 billion. It uses predictive analytics, inter alia, to obtain information on the degree of use of individual vessels.
The better repositioning of empty containers will save millions of dollars,” says Jan Voetmann of Maersk Analytics.
The problem is huge, the analysis shows that in maritime transport, a large number of containers still cover even half of their routes while empty. In the case of Maersk, the cost of transporting empty containers is a billion dollars a year.
Maersk also uses predictive analytics to predict possible failures of ships’ engines. This prevents critical situations. Incidentally, the historical analysis of the speed of ships on particular routes helped to optimise schedules so that fuel consumption dropped by several percent in three years.
Optimisation of container management is also the main effect of using predictive analytics in NileDutch, one of the leading transport companies in African markets. On one hand, the total costs of managing empty containers have been significantly reduced, while on the other, the size of the container fleet has been reduced (while maintaining the same scale of transport).
The algorithm helps to determine the number of drivers and trucks needed.
For several years now, Amazon’s logistics specialists have been able to predict where and when certain goods will be needed. This is the result of a thorough analysis of historical data, e.g. purchasing decisions, visits to websites of specific suppliers, etc. An advanced algorithm allows for calculating the demand for a given product in a given place. Amazon, therefore, ships its products to this area before the order is placed. Once the order has been placed, the lead time is really short.
In other words, when an Amazon customer orders a product, it can be shipped from a nearby centre in a much shorter time because it is available in the nearest warehouse. It also helps Amazon’s logistics to accurately predict the number of vehicles needed to operate and the drivers that will be required to perform the task at a given moment.
Predictive analytics will shorten routes and save millions of dollars.
Another example of predictive analytics comes from UPS. As a standard, this global logistics operator handles approximately 19 million parcels a day (UPS has 96,000 vehicles). It has been calculated that eliminating just one mile of each driver’s route per day could result in savings of approximately $50 million.
Predictive analytics will shorten lead times, increasing overall efficiency. The new UPS Network Planning Tools software, to be fully deployed next year in the United States, uses real-time data and analysis. It is expected to bring annual savings of up to $200 million.
DB Schenker, on the other hand, uses predictive analysis to plan processes and optimisation measures (the company has been using it for three years now). This software, used in individual warehouse locations, simulates daily processes. DB Schenker has also developed Industrial Data Space to enable secure data exchange between companies using predictive analytics, also to enable predictive maintenance (prevention of machine and vehicle failures).